FULL TEXT: CRS REPORT ON ROTH JOBS PROPOSAL.

TO: Honorable David Pryor

This proposal contains several reductions which sum to a total loss of $37.6 billion from FY93-FY98 according to estimates contained in the accompanying materials. ¹ These provisions and their respective 6-year revenue costs are: (1) prospective indexing of capital gains ($11.7 billion), (2) changes in the alternative minimum tax ($2.5 billion), (3) an increase in the limit on the option to expense equipment investment from $10,000 to $25,000 ($8.4 billion), (4) a reinstatement of fully deductible Individual retirement accounts (IRAs) including an option for backloaded accounts ($3.1 billion), (5) penalty free withdrawals of for certain purposes ($2.4 billion), (6) a temporary jobs tax credit for hiring new employees ($3.4 billion), (7) a repeal of luxury taxes ($2.6 billion), and (8) a modification of passive loss restrictions for certain individuals engaged directly in real estate activities ($2.5 billion).

There are offsetting revenue receipts from spending cuts of $45.7 billion over the 6 year period. These provisions include two changes in mandatory programs totaling $11.3 billion: elimination of the lump sum retirement benefit election for Federal civilian employees ($8.3 billion) and an administrative reform designed to reduce medicare costs (requiring essentially information reporting on whether the employee is in a group plan).

There are also a series of reductions in discretionary programs to be enforced through spending caps, which total $34.3 billion. These include reductions in Federal aid for mass transit, elimination of highway demonstration projects, an administrative provision affecting government contractors, reductions in Federal employment, reductions in administrative expenses (not specified), restrictions on accumulation of leave for senior career employees of the Federal government, elimination of the Interstate Commerce Commission, sale of Federal helium reserves, reduction of gain Services Corporations Funding, termination of the Copyright Royalty Commission, and reduction in certain foreign aid programs.

Some of these specific changes are quite small. The ones that are in excess of $1 billion include the reductions in transportation (mass transit and highway) spending ($10.5 billion), the cuts in Federal employment and unspecified cost administrative cost reductions ($19.1 billion), and foreign aid ($2 billion).

The release states that the program will increase employment by 800,000 jobs over five years, with 200,000 in the first two years. The release includes a page reporting the jobs created by the tax provisions prepared' by the minority staff of the Joint Economic Committee. We have been unable to obtain full details from the Committee on the derivation of these estimates; in the final section we discuss the materials that they did provide us. These estimates are greatly in excess of what one might expect given a standard multiplier effect, however. No estimates are presented for the offsetting contractionary effects of the spending cuts.

This memorandum will discuss first the short run effects on aggregate demand and then the long run effects on economic growth. Note that there is normally a tension between these objectives, in that a policy that reduces the deficit tend5 to be contractionary in the short run although it increases growth in the long run.

Short Run Effects Aggregate Dead

Since the revenue gains exceed the losses, the short run effects of the proposal would be expected to be contractionary -- that is, jobs would be reduced rather than gained. These effects could be characterized as negligible, however, since the net fiscal contraction is extremely small particularly in the first year or two when the concern about recovery from the recession is most serious. In 1994, the net gain is only $445 million. Thus, the proposal would be expected to have little effect on jobs.

If the capital incentives increase savings, as is suggested by the sponsors, these slight contractionary effects would be increased since an increase in savings reduces aggregate demand. There is, however, little reason to believe that the tax provisions in the proposal will increase savings because there is little evidence that increasing the rate of return increases savings. ²

Run Effects on Economic Growth

In the long run, there is no reason to expects an effect on the number of jobs even with a large change. A fiscal stimulus does not have a persistent effect on employment. Rather, the issue in the long run is the effect of the proposal on overall savings and investment.

The effect of the proposal in the long run, given the lack of evidence that tax incentives increase savings, will depend largely on the effects on the deficit. In the last year, the spending cuts approximately equal the revenue losses (the net is $95 million), which would suggest no permanent effects.

It seems likely that the proposal will reduce growth in the long run, however, because the revenue losses from the tax provisions are likely to grow substantially. Moreover, at least one of the spending cuts will eventually turn into a loss -- the elimination of lump sum Federal retirement payments. The provision is responsible for $3 billion in spending cuts in the last year estimated. Since these payments substituted for annuities, spending on annuities will rise eventually and the spending cut will becomes spending increase. ¹

One of the tax proposals, the increase in expensing for investment, will continue to decline in revenue cost. This provision loses $1 billion in the last year, and will probably become quite small. This decline can be readily seen in the revenue estimates.

Two of the tax proposals -- capital gains and IRA, -- will be likely to lose much larger sums in the future. This trend can also be seen in the revenue estimates presented.

First, the prospective capital gains provision begins at a very small revenue loss because it initially indexes only the small amount of inflation on newly purchased assets. The revenue loss grows rapidly. At 1998 levels of income, the long run4 steady state cost of capital gains indexing is estimated at about $26 billion.

Secondly, the IRA provision will grow rapidly over time given the increase in funds built up in these tax exempt accounts. We estimate this long run revenue cost to be approximately $14 billion annually at 1998 income levels.

The excess of the capital gains and IRA provisions over the amounts reported in the estimated data would be $32 billion in 1998. Netting out the $1 billion cost of the depreciation provision against the $3 billion of savings from the Federal retirement program (that will reverse sign) results in an additional cost in excess of $34 billion in 1998. This increase in the budget deficit will largely come out of private savings investment. Hence, the proposal taken as a whole would be expected to reduce overall savings and the long run level of output.

Materials Supplied by the Minority Staff of the Joint Economic Committee

The Minority staff of the Joint Economic Committee provided two documents that they indicated were relevant to the jobs estimates.

The first document was a one page summary estimating the affects of the IRA provision done by Roger Brinner of Data Resources Inc. It predicted an eventual increase of 250,000 jobs. This model is a standard short-run macroeconomic model with unemployed resources. The simulation, however, could not have been a standard simulation of the IRA provision in the proposal, since the IRA provision actually raises money in the short run. Such a straightforward simulation should have produced a contractionary effect of negligible magnitude. It appears from a footnote that the expansionary effect may reflect an assumption that individuals will withdraw and spend large amounts from IRA,, presumably because of penalty free withdrawals for certain purposes that is, that the provision will provide a reduction in saving that will be quite large. We know of no evidence to support such an assumption.

The second document is a paper entitled "Capital, Taxes and Growth", by Gary Robbins and Aldona Robbins (National Center for Policy Analysis). This paper does not provide a direct estimate of jobs for the proposal but rather outlines a model that apparently reflects some of the underlying methodology. This model is essentially a long run growth model as discussed in the previous sections and does not really address the consequences in the next few years since it has no adjustment path. This model would predict that reductions in tax burdens would increase output in the long run, because it assumes an infinitely elastic savings response. As noted above, the empirical literature does not necessarily support a savings response; even in the study where positive elasticities are found, the response is small. Because of the infinite savings elasticity, deficits do not reduce savings and investment.

² Economic theory indicates that the effects of reducing taxes on capital income has ambiguous effects on savings, due to offsetting income and substitution effects. Most time series studies of savings fail to uncover a significant relationship. See Michael Boskin, Taxation, Savings, and the Rate of Interest, Journal of Policy Economy, v. 86, January, 1978, pp. s3-s27; Barry Bosworth, Tax Incentives and Economic Growth, Washington D.C.: Brookings Institution, 1984; A. Lans Bovenberg, Tax Policy and National Savings in the United States: A Survey, Nation Tax Journal, v. 42, June, 1989, pp. pp.123-138; Irwin Friend and Joel Hasbrouck, Saving and Alter Tax Rates of Return, The Review of Economics and Statistisc, v. 65, November, 1983, pp. 537-543; E. Philip Howry and Saul H. Hymana, The Measurement and Determination of loanable Funds Savings, Smokings Papers on Economic Activity, No. 3,1978, pp. 655-705; John Makin and Kenneth A. Couch, Savings, Pension Contributions, and the Real Interest Rate, The Review of Economics and Stastics, v. 71, August, 1989, pp. 401-407. Economic theory suggests that 1AAs are not likely to increase savings because most participants are at the limit and have no tax incentive at the margin, leaving only an income effect that tends to reduce savings. Although some studies of IRAs have found a positive savings effect, those studies have been the subject of some criticism; others have found no effective. See Jane G. Gravelle, Do Individual Retirement Accounts Increase Savings? Journal of Economic Perspectives, v. 5, Spring, 1991, pp. 13-148, for a review.

³ The cutbacks in spending on mass transportation and highways would also have an effect to the extent that they reduce the stock of public capital, although these effects might not show up in measured GNP.

⁴ The current baseline is estimated at $162 billion at 1993 income levels, and indexing is estimated to result in the equivalent of a 54 percent exclusion. At current levels the revenue loss, assuming a 25.7 percent average marginal tax rate, is $22.5 billion (0.257 X $162 billion X 0.54). Based on recent research on the realizations response, we include a behavioral response that will increase realizations by about 15 percent. (See Jane G. Gravelle, Limits to Capital Gains Feedback Effects, Congressional Research Service Report 91-250, March 15,1991, and Leonard E. Burman and William C. Randolph, Measuring Permanent Responses to Capital Gains Tax Changes in Panel Data, Forthcoming, American Economic Review). The number is increased to 1998 levels to reflect a 6 percent annual nominal growth.